Mobile network operators operate and try to optimize their networks on the basis of the current network state. This network optimization is often done at the core of a mobile network from a centralized location. Information that is available at the edges of the network, such as the radio spectrum available at a location, is not used in network optimization because it is difficult to collect, and the embedded information is oftentimes lost by the time the network controller receives it.
Therefore, current network optimization and reconfiguration techniques do not take edge capacity of radio networks into account. This severely limits the possible optimizations that can be implemented since the radio network is the most congested network in current environments, and likely to remain so for the long term due to limited spectrum.
Thus, network optimization solutions which can take into account information collected at the edge of the network would be desirable.